VISIBILITY · CITATION RANK
Where does your brand rank when AI agents answer?
Citation Rank scans ChatGPT, Gemini, Perplexity, Rufus, Copilot, and Claude - and tells you exactly which queries surface you, which surface competitors instead, and what to fix first. Free scan in 24 hours.
What is Citation Rank?
Citation Rank is the position your brand occupies inside a single AI-generated answer - which place you were named in, on which engine, in which region, for one specific question. It is a per-answer measurement, not an average. Where a search rank describes your position in a list of links, Citation Rank describes your position in a sentence somebody is actually reading.
Citation Rank is your AI Share-of-Voice score across every agent surface - measured first-party, refreshed continuously, tied to the queries that drive your category.
- →Free scan: drop your URL → get your rank in 24 hours
- →Tracks 6 AI surfaces continuously after you connect your catalog
- →Reports SoV delta versus your top 8 named competitors
- →Outputs the queries to win, ranked by intent value
How it works
Four capabilities, one signal
Six platforms, one number
Your rank across ChatGPT, Gemini, Perplexity, Rufus, Copilot, and Claude - refreshed weekly, surfaced as a single comparable score.
Query-level visibility
See the top-of-funnel queries you should appear on and currently don't. We map each query to category intent so you know which gaps move revenue.
Competitor delta
Who's beating you, in which categories, with what answer the model is reaching for. Not 'how do you compare to X' - 'where is X taking the citation that should be yours.'
Prescriptive next-30-days
Three specific actions per category, ranked by expected presence lift. Not generic SEO advice - citation-anchor edits LLMs actually reward.
Why a per-answer number, and not one score
An answer engine does not return a ranked list that you can read positions off. It returns prose that names a handful of brands in an order it chose, and that order differs by engine, by region, and by how the question was phrased. Averaging those into one figure destroys the only information that tells you what to do next.
Citation Rank keeps the answer intact. Each tracked question stores the full response text, the brands named in it, the order they appeared in, and every source the engine cited to build it. When your position moves, you can open the answer that moved and read why.
This matters more than it sounds. In our own category analysis, engines cited wildly different sources for the same question - the overlap between what ChatGPT and Gemini reach for is roughly four domains in thirty. A single blended score hides exactly that.
Mentioned is not the same as recommended
A brand can be quoted as a source in an answer and never offered as an option in it. This is common and it is the single most misread signal in AI visibility: your page is good enough for the model to take a fact from, but nothing about it makes you a candidate when the reader asks who to use.
Citation Rank counts those separately. A mention records that the engine drew on you. A recommendation records that the engine put you forward. The work required to fix each is different - the first is a content and citation problem, the second is a positioning and third-party-corroboration problem.
What moves it
Three things, in rough order of effect. First, whether independent sources corroborate you - comparison pages, reviews and forum threads carry weight your own copy cannot, because they are not yours. Second, whether your own pages state checkable things: a short declarative sentence that defines something is what models retrieve, far more often than a statistic. Third, whether your product data is unambiguous enough for an engine to describe you confidently rather than reaching for a competitor it understands better.
None of those move in a day. Citation Rank is designed to be read weekly, against the specific questions your buyers ask, rather than watched hourly.
Case study
From #6 to #1 across 3 LLMs
Side by side
Citation Rank, Share of Voice and Visibility Score - which one answers your question
These three measure adjacent things and buyers routinely conflate them. The short version: Citation Rank is per-question, Share of Voice is per-category, Visibility Score is the roll-up. Most teams need the first two and report the third.
| Citation Rank · | Share of Voice | Visibility Score | |
|---|---|---|---|
| Question it answers | For this specific prompt, where did we appear? | Across the category, how much of the answer space is ours? | Overall, are we getting more or less visible? |
| Unit of measurement | One prompt, one engine, one region | A cluster of prompts, aggregated | All tracked prompts, rolled into one figure |
| Changes when | A single answer reorders | Competitors gain or lose ground across a topic | The aggregate moves - usually slowly |
| Best for | Diagnosing why a specific answer went wrong | Deciding which topics to invest in next | Reporting the trend to people who will not read prompt-level data |
| Who uses it | SEO and content leads | Category and growth leads | Executives and board reporting |
| Fails when | Used alone - one prompt is noisy | Read without knowing which prompts are in the cluster | Treated as a diagnosis rather than a summary |
| Mentions vs recommendations | Separated per answer | Separated per cluster | Blended - by design |
A single composite score cannot tell you what to fix. If a number moves and you cannot name the answer that moved it, you are looking at the wrong one of these three.
Go deeper
Related products and proof
Share of Voice
Zoom out from one answer to a whole category.
Visibility Score
The roll-up figure, and why a single score has limits.
Cited, recommended, or neither
Why a citation and a recommendation are different numbers and should never be blended.
Agentic commerce statistics 2026
The first-party corpus these numbers come from: 1,795 buyer questions, five engines, weekly.
FAQ
Common questions
Yes, and the engine matters far more than the country. In our September 2026 category analysis, two engines answering the same question shared as few as four of their top thirty cited domains, while the same engine across three countries shared twenty-four of thirty. Track per engine first.
Weekly is the useful cadence for most brands. The inputs that move it - third-party corroboration, content structure, product data quality - do not change daily, and checking more often mostly surfaces normal answer variance rather than a real shift.
Google ranking measures your position in a list of links. Citation Rank measures your position inside a generated answer - which brand was named first, second, third, and whether you were named at all. The mechanisms differ too: the citation graph, training data and structured product data decide an answer, not classic SERP signals. A brand can rank first on Google and be absent from the answer an assistant gives to the same question.
ChatGPT, Gemini, Perplexity, Rufus, Copilot, and Claude. We add new surfaces as their commerce queries become trackable.
Weekly for active customers; daily for Enterprise. The free 24-hour scan is a point-in-time snapshot.
No - the free scan only needs your URL. Catalog connection unlocks the prescriptive recommendations and the continuous refresh.